Clone or Download project and import "sound.py" to your code:
import soundImport Noise class from "sound.py" and initial it with a Wav file or byte array of a sound:
from sound import Noise
noise1 = Noise(path='./M1.wav', name='Mic 1')
# plot your noise
noise1.draw()| Method | Usage |
|---|---|
| parse_wav() | parse wav file to byte array |
| init_fft() | initials fft of noise |
| get_fft() | returns fft of noise |
| get_energy() | returns energy (power) of noise |
| get_duration() | returns duration of noise in seconds |
| get_slice() | returns slice of noise like 2 seconds of noise |
| get_max() | returns upper bound of amplitude |
| when_max() | returns when upper bound happends in seconds |
| shift_right() | shifts noise signal to right in time domain |
| shift_left() | shifts noise signal to left in time domain |
| reverse() | reverses noise |
| convolve(other) | convolve the noise to another or use '*' operator like 'n1 * n2' |
| draw() | plot the noise in time domain |
| draw_fft() | plot the noise fft in frequency domain |
| draw_all() | plot the noise and fft both |
| draw_phase() | plot phase of voise in frequency domain |
| hear_noise() | returns IPython.display.audio object to hear noise in Jupyter Notebook |
- matplotlib >3.3.0
- numpy >1.19.1
- scipy >1.5.1
- ipython >7.16.1
- For better experince install "Jupyter Notebook"